Volume 4 ~ Issue 12 ~ Version 4 (December 2014)

1. IOSR Journal of Engineering (IOSRJEN) www.iosrjen.org
ISSN (e): 2250-3021, ISSN (p): 2278-8719
Vol. 04, Issue 12 (December 2014), ||V4|| PP 35-40
International organization of Scientific Research 35 | P a g e
Optimization of the Fermentation Medium to Receive The
Highest Biomass Yield By Bacillus Subtilis Natto And The Initial
Test Of Nattokinase Yield
Tuan, Nguyen Anh1
; Huong, Nguyen Thuy1
1
Department of Biotechnology, Ho Chi Minh City University of Technology, Vietnam
ABSTRACT: A bacterium, named Bacillus subtilis natto, is employed in medium optimization in order to be
applied for nattokinase enzyme production. Six factors including glucose (X1), soybean peptone (X2), K2HPO4
(X3), MgSO4.7H2O (X4), NaCl (X5) and CaCl2 (X6) were used to screen some effect factors in the fermentation
medium for the production of biomass by Plackett – Burman design. In the range studied, soybean peptone and
CaCl2 had significant effect on biomass production. The optimized medium containing (g/L) glucose: 5.625,
soybean peptone: 13, K2HPO4: 2.125, MgSO4.7H2O: 0.875, NaCl: 5, CaCl2: 0.05 were used for production of
biomass and the highest dried cell weight in broth was 3.033 g/L. The maximum nattokinase yield was 31.06 ±
0.297 FU/mL of substrate in optimized medium above which was higher than that in before by over 30% after 20
hours of fermentation in 37o
C, pH 7.5.
KEYWORDS: Bacillus subtilis, Bacillus subtilis natto, extracellular proteases, Nattokinase, Plackett - Burman
I. INTRODUCTION
A novel fibrinolytic enzyme from natto was introduced by Sumi (1987) and it was named nattokinase.
It is one of the most considerable extracellular enzymes produced by Bacillus subtilis natto fermentation [1].
Besides Bacillus subtilis, alternative sources of obtaining this enzyme are Pseudomonas sp. and marine
creatures. Moreover, the enzyme is also obtained from other traditional fermented foods such as Chinese
douche, Korean doen-jang, Korean Chungkook-jang soy sauce. The enzyme is due to its direct fibrinolytic
activity. It could be used by the oral thrombolytic therapy or via natto food [2]. As a functional food, nattokinase
owns several benefits compared to the available clinical thrombolytic drugs, such as safety, low cost, confirmed
efficacy, prolonged effects, preventative use, and easy oral administration [3]. Several Bacillus species have
been found to be strongly associated with these fermented –soybean products. They can be grown over a wide
pH range (with an active growth between pH 5.5 – 8.5) and produce several enzymes (for example, proteases) as
well as other useful biological compounds which bring superiority for Bacillus species in the soybean
fermentation. In this paper, the used bacterium was isolated from Vietnamese natto foods. It was identified by
16S rRNA and checked some initial tests. Because of extracellular enzymes depending on the number of spores
in broth, as meaning, the level of nattokinase is high when the weight of biomass is large. That is the reason, we
determine the parameter of fermentation medium by Bacillus subtilis natto to accumulate the large biomass in
the broth [4].
Many comprehensive studies were done to determine parameters of the optimal fermentation
conditions for nattokinase production. Essential nutrients are critical for enhancing nattokinase production, such
as soybean peptone. Bioprocess technologies involve multiple parameters to adjust and complication so that
they need effective problem solving methods. Bioprocess technologies are complicated processes which involve
multiple parameters to adjust suitable condition of fermentation. An accurate mathematical model equation in
order to describe the whole process, interactions among of variables, statistically designed experiments use a
small set of carefully planned experiments. This method is more satisfactory and effectively than other methods
such as classical one – at – a – time or mathematical methods because it can study many variables
simultaneously with a low number of observations, save time and costs. The components in fermentation
medium are employed to optimize batch culture for biomass production. They are designed by Plackett-Burman
matrix, six factors with 12 runs. The replication per factor is enough large to analyze ANOVA. In all of these
cases, a mathematical model was used to describe the relationship between biomass and the concentration of the
all factors. The response surface methodology (RSM) is an effective strategy for optimizing biomass production
by Bacillus subtilisnatto in a shaken flask culture [2, 5]. The present work describes the successful optimization
of a culture medium for the hyper-production of biomass by Bacillus subtilis natto to apply in the production of
nattokinase.

2. Optimization Of The Fermentation Medium…
International organization of Scientific Research 36 | P a g e
II. THE MATERIALS AND METHODS
Materials, medium and culture conditions
The micro-organism species: Bacillus subtilis natto was isolated from Vietnamese natto-food
The cultural medium: Bacillus subtilis natto grows in the nutrition broth (NB) and nutrition agar (NA), pH
7.5, the temperature at 37o
C, the agitation rate of 150 rpm. After 20 fermentation hours, the cell rate in the broth
was 50 billion colony-forming units. The fermentation medium (g/L): glucose (1.25 – 10.00), soybean peptone
(5 -15), K2HPO4 (1.25 – 3.00), MgSO4.7H2O (0.25 – 1.50), NaCl (2.5 – 7.5) and CaCl2 (0.05 – 0.40) were
employed in the range studied.
The fermentation condition: temperature 37o
C, initial pH in medium 7.5 ± 0.3, adjusted by NaOH 1N solution
and the agitation rate of 150 rpm.
Experiment designs
Checking some initial tests of Bacillus subtilisnatto
 Observing the colonies of Bacillus subtilis natto.
 Observing micro-cell under microscope 100X.
 Bacillus subtilisnatto was identified by 16S rRNA.
 Checking nattokinase activity.
Optimizing the fermentation medium for production of biomass by Bacillus subtilis natto : Bacilus subtilis
natto was incubated in NB medium. After 20 inoculum-hours, the number of the cells was checked and counted
as well as added approximately 5 billion colony-forming units per milliliter of fermentation medium.Six
variables were examined in the Plackett-Burman matrix with different 12-run. We determined the dried cell
weight for each validation formula and analyzed the factors that affect the biomass production by Bacillus
subtilis natto. The main factors in experiments had p-value < 0.05. With the selected factors, we carried out the
first experiments with the original values (+1, -1). After analyzing the initial experiments, we determined
whether the factors having great impacts on the high regression equation suitably or not. Based on that, we
conducted the experiments for response surface methodology having the central composite designs (RSM-CCD)
and determined function of the polynomial regression accurately to describe relations between the dried cell
weight in broth and medium factors. The results of response optimizer that were simulated by software were
measured by experiments to determine the highest actual biomass in broth. Furthermore, the level of nattokinase
in broth was analyzed and compared with the initial experiments in NB medium.
The analyzing method
Analyzing biomass in fermentation broth: Broth was centrifuged at 4000 rpm for 10 minutes, removed the
supernatant above, biomass was obtained and dried at 1050
C for 5 minutes. After that the dried cell weight was
measured.
Analyzing nattokinase in broth: Nattokinase activity was determined by the ability to hydrolyze fibrin fibers.
Bacillus subtilisnatto was stopped the fermentation after 20 hours and broth was centrifuged at 13 000 rpm for
20 minutes and obtained the supernatant to determine nattokinase activity. Tris–HCl (50 mM, pH 7.5) of 1.3 mL
and 0.4 mL of 0.72% (w/v) fibrinogen solution were taken in vials and kept in water bath (37o
C) for 5 minutes.
Then 0.1 mL thrombin (20 U/mL) was added and kept in water bath (37o
C) for 10 minutes. To this clot, 0.l mL
of enzyme was added. After incubation (37o
C, 60 minutes), 2 mL of 0.2 M trichloroacetic acid (TCA) was
added. Vials were kept 20 minutes and centrifuged at 3000 x g for 5 minutes. One unit enzyme activity is
defined as the amount enzyme required to produce an increase in absorbance equal to 0.01 in 60 minutes at 280
nm [6].
III. RESULT AND DISCUSSION
Some initial tests of Bacillus subtilisnatto : A bacterium that was isolated from Vietnamese natto food, is
named Bacillus subtilis natto.

3. Optimization Of The Fermentation Medium…
International organization of Scientific Research 37 | P a g e
Figure 1. Bacillus subtilis natto under microscopes and its colony on blood agar
It is rod-shaped, oval spores, Gram-positive and the colony size is approximately 0.5- 0.8 µm x 1.5- 3.0 µm.
The colonies are round, their edges are irregular jagged, their diameters are from 3 to 5 mm, and they have light
yellow, dark heart. After 24- hour culture they become wrinkled surface. The bacterium that was employed to
study had the nattokinase yield of 23.583 ± 1.539 FU/mL of substrate in NB medium.Various culture-
independent methods have been developed; in particular, methods using the variable and conserved regions of
the16S rRNA have proved successful in characterizing the gut microbiota. Sequencing of 16S rRNA genes has
revealed that microbial diversity in the gut is far more extensive than previously described from studies of
cultured microorganisms alone. And the result of 16S rRNA was proved that the bacterium that the work was
employed is named Bacillus subtis. Based on the result, we study the optimized fermentation medium to be
obtained the maximum biomass yield.
Optimizing fermentation medium for biomass production by Bacillus subtilisnatto
Screening main effective factors of the biomass production by Bacillus subtilis : The results of screening
experiments were interested in the main effect factors. The efficiency of the selecting experiments and the
degree of these factors’ impact in the fluctuation range according to different levels were shown in Table … The
obtained dried-cell weight was noticeably depended on two factors, i.e. soybean peptone (X2) and calcium
chloride (X6) (p-value < 0.05). The polynomial regression (Rsq is 86.7%) is determined according to the simple
function as given below:
Response, Ys = 1.265 + 0.0539 X2 - 1.004 X6 (3.1)
In the above equation, y (g/L) symbols for dried cell weight, x2 (g/L) symbols for added soybean
peptone and x6 (g/L) symbols for calcium chloride
Table 1. The factors in Plackett-Burman matrix and its effects on biomass production by Bacillus subtilis
Name of factors, g/L
Symbols of
factors
Values of factors Main effect
(R-sq = 86.7%)
p-value
Low
(-1)
High
(+1)
Glucose X1 1.25 10.00 +0.164 0.266
Soybean peptone X2 5.00 15.00 +0.539 0.009
K2HPO4 X3 1.25 3.00 +0.164 0.266
MgSO4.7H2O X4 0.25 1.50 +0.269 0.095
NaCl X5 2.50 7.50 -0.141 0.333
CaCl2 X6 0.05 0.40 -0.351 0.044
The optimal values of parameters in fermentation medium to get maximum the dried cell weight in broth
We went on conducting 9 experiments, 4 of them are (-1,1) ones and 5 are the central ones. ANOVA
was carried out to statistically analyze the correlation of biomass (dried cell weight) to the two selected factors.
The p-value of Lack-of-fit test was lower than 0.001 and R-sq was 90.56 %.

4. Optimization Of The Fermentation Medium…
International organization of Scientific Research 38 | P a g e
Table 2. Dried-cell weight in fermentation broth by RSM-CCD experiments.
RunOrder
Coded units Uncoded units Response, Y_CCD (dried
cell weight, g/L)X2 X6 X2 (g/L) X6 (g/L)
1 -1 -1 5 0.05 2.826
2 0 0 10 0.225 2.791
3 0 0 10 0.225 2.791
4 0 0 10 0.225 2.791
5 1 -1 15 0.050 3.072
6 1 1 15 0.400 3.002
7 -1 0 5 0.225 2.580
8 -1 1 5 0.400 2.475
9 1 0 15 0.225 2.861
10 0 0 10 0.225 2.756
11 0 1 10 0.400 2.897
12 0 0 10 0.225 2.756
13 0 -1 10 0.050 3.037
This means that the arrangement of the two big effect factors (Table 2) is near the extreme of the aimed
function and the polynomial regression between two selected factors and response (dried cell weight) could be
given in the function of the poly-nominal regression. At this, the highest yield of all experiments was obtained.
To establish the right relationship between X2 and X6, we analyzed the 13 RMS-CCD experiments (shown in
Table 2). The response that dried cell weight (g/L) in the broth ranged from 2.475 to 3.037. In the eighth-run
(soybean peptone 5g/L and CaCl2 0.4 g/L), the dried cell weight was obtained lower than 2.5 g/L. Most
experiments were given that the dried cell weight in the broth was higher than 3.0 g/L and those were designed
with CaCl2 0.05g/L and over 5 g/L of soybean peptone.
The polynomial regression equation (R-sq = 97.08%) was shown:
Response, YCCD (g/L) = 2.643 + 0.088 X2 - 3.652 X6 - 0.004 X2
2
+ 5.142 X6
2
+ 0.08 X2.X6 (3.2)
In the above equation, y (g/L) symbols for dried cell weight, x2 (g/L) symbols for added soybean peptone and
x6 (g/L) symbols for calcium chloride.
The statistical significance of the model equation was evaluated by F-test for analysis of variance
(ANOVA), which showed that the regression is statistically at 99% (p < 0.0001) confidence level. The model
F-value of 46.47 for biomass production implies that the model is statistically significant. The coefficient of
determination (R2
) was calculated to be 97%, indicating that the model could explain 97% of the variability.
The “lack of fit test” compares the residual error to the “Pure Error” from replicated design points. The
estimated models fit the experimental data adequately. The regression equation was given that the dried cell
weight in the fermentation environment was the high number of all when the concentration of soybean peptone
was increased (+0.088) and the concentration of calcium chloride was decreased (-3.652). The graph below
showed that the relationship between the two factors soy peptone, calcium chloride with dry biomass of the
broth when hyper-production of biomass by Bacillus subtilis was conducted in optimal condition and medium
from model 3.2.
X2
X6
15.012.510.07.55.0
0.40
0.35
0.30
0.25
0.20
0.15
0.10
0.05
>
–
–
–
–
–
< 2.5
2.5 2.6
2.6 2.7
2.7 2.8
2.8 2.9
2.9 3.0
3.0
Y_CCD
Contour Plot of Y_CCD vs X6, X2
5
01
42.
6.2
2.8
4.0
0.3
0 2.
1.0
15
03.
DCC_Y
6X
2X
2X,6XsvDCC_YfotolPecafruS
Figure 2. Contour plot and surface plot of response in RSM - CCD
From the response surface and contour plot (Fig.2) it is obvious that soybean peptone and CaCl2 had a
significant effect on biomass production compared to other variables. The biomass yield increased if the metal
ions such as Ca2+
decreased, because of the enzyme activation in biochemical reactions by the metal ions.

5. Optimization Of The Fermentation Medium…
International organization of Scientific Research 39 | P a g e
Sumi et al. (1987) reported that the presence of a trong fibrinolytic enzyme in natto, and nattokinase
may be an equally good protease for oral fibrinolytic therapy because of its confirmed safety for long-term
intake, stability and the strong fibrinolytic activity. In addition, nattkinase is extracellular enzyme, the level of
it depends on biomass yield. If the biomass yield is maximum, the level of nattokinase is high [1, 5].The
fermented soybeans produced using bacteria, however, may have a preventive effect not only on thrombosis
but also cancer. Statistical experimental design has not been widely used in the biological sciences even though
it has been commonly employed in many other areas such as industrial, chemical, engineering, agricultural,
medical, and food sciences. The primary reason for this is that most biological research has not been involved
in many manufacturing processes. However, since genetic engineering, biomaterials, and bioprocess
technologies like biodegradation and bioremediation have emerged, more people are getting interested in
experimental designs to improve their biological processes and productions by shortening time and increasing
efficiencies [7]. The fermentation media for biomass production was optimized using statistical method.
According to the previous report using the medium components such as soybean peptone, calcium chloride
resulted in maximum biomass. In addition to establishing optimal fermentation medium compositions, the
present methodology also makes it possible to predict the yield if the composition of the medium is altered in
some ways, by using the quadratic equation. Central composite experimental design maximizes the amount of
information that can be obtained, while limiting the numbers of individual experiments required. Thus small
and less time consuming experimental designs could generally suffice for the optimization of many
fermentation processes. Therefore, with the increase in yield and productivity and simultaneous cost reduction,
the industrial biomass production by B. subtilis can be regarded as possible and economically attractive.
Comparison biomass and the level of nattokinase between the initial experiments and experiments in
optimum medium : The biomass production was 2.777 ± 0.0192 g/L in the original medium with six
components at their central levels. The maximum biomass yield was 3.033 ± 0.243 g/L in optimized medium
composed of (g/L): glucose (5.625); soybean peptone (13), K2HPO4 (2.125); MgSO4.7H2O (0.875); NaCl (5);
CaCl2 (0.05) which was higher than the initial medium by 10%. Additionally, the nattokinase yield was 23.583
± 1.539 FU/mL of substrate in NB medium. The maximum nattokinase yield was 31.06 ± 0.297 FU/mL of
substrate in optimized medium above which was higher than that in before by over 30%.
The present of calcium and magnesium ions in culture medium is important in cell viability. Calcium
ions take part in synergistic interactions with enzymes responsible for anchoring surface proteins to the cell
wall, thereby affecting the bacterium’s adhesion ability [8]. Magnesium ions play a role in peptidoglycan
synthesis, cell wall strength, and the prevention of cell lysis. Previous studies on the metal binding behavior of
Bacillus subtilis have focused on the metal binding capacity and affinity Calcium and magnesium ions are both
important biologically active metal ions that are some of the most abundant divalent cations in nature. We find
that electrostatic effects are responsible for a strong binding between metal ions. Those could lead the change in
density and weight of Bacillus subtilis as well as the level of nattokinase in broth to be increased. The
hypothesis cannot be proved because of lacking of data which is collected from lab-work. However these data
could be able to explain how the increase of nattokinase yield and biomass production is.
IV. CONCLUSION
The maximum biomass yield was 3.033 ± 0.243 g/L in optimized medium composed of (g/L): glucose
(5.625); soybean peptone (13), K2HPO4 (2.125); MgSO4.7H2O (0.875); NaCl 5; CaCl2 (0.05) which was higher
than the initial medium by 10%. Additionally, the maximum nattokinase yield was 31.06 ± 0.297 FU/mL of
substrate in optimized medium above which was higher than that in before by over 30%.
V. ACKNOWLEDGEMENTS
We would like to fully appreciate the help of my close-friend, named Tam, Tran Thi Minh in the
Chemistry Faculty of Ho Chi Minh City University of Technology in this study. She helped us to design the
experiments and check the papers. Moreover, we thank the Nam Khoa company for identification of the bacteria
name.
REFERENCES
[1] Dabbagh, F., et al., Nattokinase: production and application. Applied microbiology and biotechnology, 2014: p. 1-8.
[2] Berenjian, A., et al., Nattokinase production: Medium components and feeding strategy studies. Chemical Industry and Chemical
Engineering Quarterly, 2014(00): p. 37-37.
[3] Trần Quốc Tuấn, N.T.T.K., Lê Thị Thúy Ái, Đinh Minh Hiệp, Trần Cát Đông, Optimization of medium composition for
recombinant Nattokinase production from bacillus subtillis by using response Surface methodology. Biology Journal, 2014.
36(1SE): p. 130 - 137.
[4] Lê Thị Bích Phượng, V.T.H., Trần Thạnh Phong, Lê Tấn Hưng, Trương Thị Hồng Vân, Lê Thị Hương, Isolation and selection of
some Bacillus strains capable of nattokinase production. Biology Journal, 2012. 34(3SE): p. 99 - 104.

6. Optimization Of The Fermentation Medium…
International organization of Scientific Research 40 | P a g e
[5] Deepak, V., et al., Optimization of media composition for Nattokinase production by Bacillus subtilis using response surface
methodology. Bioresource Technology, 2008. 99(17): p. 8170-8174.
[6] Cho Y. H., S.J.Y., Kim K.M., Lee I.Y., Kim S.B., Kim H.S., Han N.S., Lee B.H., Kim B.S., Production of nattokinase by batch
and fed-batch culture of Bacillus subtilis. New Biotechnology, 2010. 27(4): p. 341 - 346.
[7] Kwon, E.-Y., et al., Production of nattokinase by high cell density fed-batch culture of Bacillus subtilis. Bioprocess and biosystems
engineering, 2011. 34(7): p. 789-793.
[8] Thomas III, K.J. and C.V. Rice, Revised model of calcium and magnesium binding to the bacterial cell wall. BioMetals, 2014: p. 1-
10.